Food frequency questionnaires (FFQ) are increasingly used in epidemiological study of diets because of their ease of application over prolonged assessment periods. While many such studies have assessed the validity of FFQ for adult patients, fewer apply to children and adolescents. It is well known that the first 2o years of life are the time good nutrition habits are acquired that may later prevent the development of major chronic diseases. Therefore youth and teenagers should be identified as groups that need more attention. To meet both nutrient research and maintenance needs we must have a reliable instrument that accurately assess what children and adolescents are eating. The purpose of this pilot project is to assess and validate a semi- quantitative FFQ previously by us in a sample of Puerto Rican school children. To accomplish this our specific aims are: (1) Administer the FFQ and verify its reproducibility by comparing the results to those obtained in a second administration. (2) Validate the FFQ by comparing foods listed in it to foods listed in food records that will be compiled by students for 14 days. (3) Select approximate 120 students each in two age brackets- (i) children attending 4th and 5th grade and hence are capable of understanding and completing FFQ's and food records, (ii) teenagers in the 10th and 11th grades whose acting habits are quite different that those of young children. (4) Determine the accuracy of food records through daily monitoring of lunch records by trained observers. Once satisfactory validation has been made, this approach could be useful in other school based settings or in commonly studies of diet, lifestyle and health as well as serve as a basis for larger scale dietary studies by ourselves and health researchers or governmental agencies.